icuSources/test/testdata/break_rules/line_normal.txt

   1 #
   2 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
   3 #
   4 # file: line_normal.txt
   5 #
   6 # Reference Line Break rules for intltest rbbi/RBBIMonkeyTest
   7 #
   8 # Note: Rule syntax and the monkey test itself are still a work in progress.
   9 #       They are expected to change with review and the addition of support for rule tailoring.
  10 #
  11 #         Line Breaking Rules
  12 #         Implement default line breaking as defined by
  13 #         Unicode Standard Annex #14 Revision 34 for Unicode 8.0
  14 #         http://www.unicode.org/reports/tr14/
  15 #         tailored as noted in 2nd paragraph below..
  16 #
  17 #         TODO:  Rule LB 8 remains as it was in Unicode 5.2
  18 #         This is only because of a limitation of ICU break engine implementation,
  19 #         not because the older behavior is desirable.
  20 #
  21 #         This tailors the line break behavior to correspond to CSS
  22 #         line-break=normal (BCP47 -u-lb-normal) as defined for languages other than
  23 #         Chinese & Japanese.
  24 #         It sets characters of class CJ to behave like ID.
  25
  26
  27 type = line;
  28 locale = en@lb=normal;
  29
  30 AI = [[:LineBreak =  Ambiguous:] - [\u2640\u2642]];
  31 AL = [[:LineBreak =  Alphabetic:]-[\u2764]];
  32 BA = [:LineBreak =  Break_After:];
  33 BB = [:LineBreak =  Break_Before:];
  34 BK = [:LineBreak =  Mandatory_Break:];
  35 B2 = [:LineBreak =  Break_Both:];
  36 CB = [:LineBreak =  Contingent_Break:];
  37 CJ = [:LineBreak =  Conditional_Japanese_Starter:];
  38 CL = [:LineBreak =  Close_Punctuation:];
  39 CM = [[:LineBreak =  Combining_Mark:] \u200D];
  40 CP = [:LineBreak =  Close_Parenthesis:];
  41 CR = [:LineBreak =  Carriage_Return:];
  42
  43 EB = [\u261D\u26F9\u270A-\u270D\U0001F385\U0001F3C3-\U0001F3C4\U0001F3CA-\U0001F3CB\U0001F442-\U0001F443\U0001F446-\U0001F450\U0001F466-\U0001F469\U0001F46E\U0001F470-\U0001F478\U0001F47C\U0001F481-\U0001F483\U0001F485-\U0001F487\U0001F4AA\U0001F575\U0001F590\U0001F595-\U0001F596\U0001F645-\U0001F647\U0001F64B-\U0001F64F\U0001F6A3\U0001F6B4-\U0001F6B6\U0001F6C0\U0001F918];
  44 EM = [\U0001F3FB-\U0001F3FF];
  45
  46 EX = [:LineBreak =  Exclamation:];
  47 GL = [:LineBreak =  Glue:];
  48 HL = [:LineBreak =  Hebrew_Letter:];
  49 HY = [:LineBreak =  Hyphen:];
  50 H2 = [:LineBreak =  H2:];
  51 H3 = [:LineBreak =  H3:];
  52 ID = [[:LineBreak =  Ideographic:] CJ [\u2640\u2642\u2764]];
  53 IN = [:LineBreak =  Inseperable:];
  54 IS = [:LineBreak =  Infix_Numeric:];
  55 JL = [:LineBreak =  JL:];
  56 JV = [:LineBreak =  JV:];
  57 JT = [:LineBreak =  JT:];
  58 LF = [:LineBreak =  Line_Feed:];
  59 NL = [:LineBreak =  Next_Line:];
  60 NS = [:LineBreak =  Nonstarter:];
  61 NU = [:LineBreak =  Numeric:];
  62 OP = [:LineBreak =  Open_Punctuation:];
  63 PO = [:LineBreak =  Postfix_Numeric:];
  64 PR = [:LineBreak =  Prefix_Numeric:];
  65 QU = [:LineBreak =  Quotation:];
  66 RI = [:LineBreak =  Regional_Indicator:];
  67 SA = [:LineBreak =  Complex_Context:];
  68 SG = [:LineBreak =  Surrogate:];
  69 SP = [:LineBreak =  Space:];
  70 SY = [:LineBreak =  Break_Symbols:];
  71 WJ = [:LineBreak =  Word_Joiner:];
  72 XX = [:LineBreak =  Unknown:];
  73 ZW = [:LineBreak =  ZWSpace:];
  74 ZJ = [\u200D];
  75
  76 # TODO: adjustment to sets needed only until Unicode properties are updated for Emoji.
  77 ID   = [ID - EB];
  78 AL   = [AL - EM];
  79
  80 dictionary = [:LineBreak = Complex_Context:];
  81
  82 # Redfine AL. LB1.  TODO: refine according to latest UAX.
  83 AL = [ AL AI SA SG XX ];
  84
  85 LB4:        BK ÷;
  86 LB5:        CR LF;
  87 LB5.1:      CR ÷;
  88 LB5.2:      LF ÷;
  89 LB5.3:      NL ÷;
  90
  91 LB6:        . (BK | CR | LF | NL);
  92 LB6.1:      [^BK CR LF NL SP ZW] CM* (BK | CR | LF | NL);
  93
  94 # Rules LB14 - LB17.
  95 # Moved before LB7, because they can match a longer sequence that would also match LB7,
  96 # for example, the sequence "OP CM SP AL" matches LB14 while the prefix of it,
  97 #                           "while only the prefix "OP CM SP" matches LB7.1
  98 LB14:        OP CM* SP* .;
  99 LB15:        QU CM* SP* OP;
 100 LB16:        (CL | CP)CM* SP* NS;
 101 LB17:        B2 CM* SP* B2;
 102
 103 LB7.1:      [^ZW SP] CM* [SP ZW];
 104 LB7.2:      [ZW SP] [SP ZW];
 105
 106 # LB8, ICU differs from UAX-14,
 107 #    ICU:    ZW ÷;
 108 #    UAX 14: ZW SP* ÷;
 109 LB8:        ZW ÷;
 110
 111 # LB8a, from Emoji proposal L2/16-011R3
 112 #      ZWJ x ID
 113 LB8a:       ZJ (ID | EB | EM);
 114
 115
 116 # LB9:  X CM -> X
 117 # LB10: Unattached CM -> AL
 118
 119 #LB11:       × WJ;
 120 #            WJ ×
 121
 122 LB11.1:      [^BK CR LF NL SP ZW] CM* WJ;
 123 LB11.2:      SP WJ;
 124 LB11.3:      WJ CM* [^CM];
 125
 126 LB12:        GL CM* [^CM];
 127
 128 LB12a:       [^SP BA HY] CM* GL;
 129
 130 # LB 13 ICU Tailoring, matches tailoring exmaple 8 from UAX 14.
 131 #
 132 #   LB13.1   [^SP] CM* [CL CP EX IS SY]    # original UAX 14 rule.
 133 #   LB13.2   SP    CM* [CL CP EX IS SY]
 134
 135 LB13.1: [^NU SP] CM* [CL CP IS SY];
 136 LB13.2: [^SP] CM* EX;
 137 LB13.2: SP [CL CP EX IS SY];
 138
 139
 140 # LB 14-17 are moved above LB 7.
 141
 142 LB18:        SP ÷;
 143
 144 LB19:        . CM* QU;
 145 LB19.1:      QU CM* [^CM];
 146
 147 # LB 20   Break before and after CB.
 148 #         Interaction with LB8a:  ZJ x ID is tricky because CM includes ZJ.
 149 #                                 ZJ acts like a CM to the left, combining with CB.
 150 #                                 ZJ acts independently to the right, no break from ID by LB8a.
 151 LB20:        . CM* ÷ CB;
 152 LB20.1a:     CB CM* ZJ (ID | EB | EM);
 153 LB20.1b:      CB CM* ÷;
 154
 155 # Note: Rule 21a must come before 21 to prevent 21.1 from matching HL BA, then
 156 #       not picking up the continuing match after the BA from 21a.
 157 LB21a:       HL CM* (HY | BA) CM* [^CM CB];
 158
 159 LB21.1:      . CM* [BA HY NS];
 160 LB21.2:      BB CM* [^CM CB];
 161
 162 LB21b:       SY CM* HL;
 163
 164 LB22.1:        (AL | HL | CM) CM* IN;   # The CM is from LB10, treat an unattached CM as AL.
 165 LB22.2:       EX CM* IN;
 166 LB22.3:       (ID | EB | EM) CM* IN;
 167 LB22.4:       IN CM* IN;
 168 LB22.5:       NU CM* IN;
 169
 170 LB23.1:      (ID | EB | EM) CM* PO;
 171 LB23.2:      (AL | HL | CM) CM* NU;
 172 LB23.3:      NU CM* (AL | HL);
 173
 174 LB24.1:      PR CM* (ID | EB | EM);
 175 LB24.2:      PR CM* (AL | HL);
 176 LB24.3:      PO CM* (AL | HL);
 177 LB24.4:      (AL | HL | CM) CM* (PR | PO);
 178
 179 # Numbers. Equivalent to Tailoring example 8 from UAx 14.
 180 LB25:        ((PR | PO)CM*)? ((OP | HY)CM*)? NU (CM*(NU | SY | IS))* (CM*(CL | CP))? (CM*(PR | PO))?;
 181
 182 LB26.1:      JL CM* (JL | JV | H2 | H3);
 183 LB26.2:      (JV | H2) CM* (JV | JT);
 184 LB26.3:      (JT | H3) CM* JT;
 185
 186 LB27.1:      (JL | JV | JT | H2 | H3) CM* IN;
 187 LB27.2:      (JL | JV | JT | H2 | H3) CM* PO;
 188 LB27.3:      PR CM* (JL | JV | JT | H2 | H3);
 189
 190 # LB28 Do not break between Alphabetics.
 191 #      Unattached (leading) CM treated as AL.
 192 LB28:        (AL | HL | CM)CM* (AL | HL);
 193
 194 LB29:        IS CM* (AL | HL);
 195
 196 # LB30  is adjusted for unattached leading CM being treated as AL.
 197 LB30.1:      (AL | CM | HL | NU) CM* OP;
 198 LB30.2:      CP CM* (AL | HL | NU);
 199
 200 # LB31  keep pairs of RI together.
 201 LB30a.1:     RI CM* RI CM* [BK CR LF NL SP ZW WJ GL CL CP EX IS SY QU BA HY NS];
 202 LB30a.2:     RI CM* RI CM* ZJ (ID | EB | EM);
 203 LB30a.3:     RI CM* RI CM* ÷;
 204
 205 # LB30b Do not break between Emoji Base and Emoji Modifier
 206 LB30b:       EB CM* EM;
 207
 208 # LB31 Break Everywhere Else.
 209 #      Include combining marks
 210 LB31.1:        . CM* ZJ (ID | EB | EM);
 211 LB31.2:        . CM* ÷;